News

⑴ Russian central bank officials stated that, despite the intensified domestic gasoline shortage following drone attacks on refining facilities and the beginning of gasoline imports by sea from countries such as India to ease supply pressure, this factor is not expected to create significant depreciation pressure on the ruble, as increased crude oil exports will provide a hedging effect on the trade front.⑵ Russia has currently launched a monthly gasoline import plan of approximately 400,000 tons, with related cost estimates at around 600 million rubles. Meanwhile, queues at domestic gas stations remain prominent. From a foreign exchange mechanism perspective, increased imports usually mean higher demand for foreign currency, but the central bank believes that, since part of the crude oil cannot be processed domestically and is instead exported, the overall change in trade flow is limited and will not significantly alter the supply and demand structure for foreign exchange.⑶ Market data shows that the ruble has depreciated by about 10% since early June, but recently experienced a short-term rebound due to central bank interest rate policy signals. Analysts pointed out that the current ruble trend is more influenced by distortions in trade structure and the pace of energy exports; while the fuel crisis brings marginal pressure, as long as crude oil exports remain resilient, there is still some buffer space against sustained impacts on the exchange rate.

BlockBeats News, July 2, according to Hyperinsight monitoring, in the past 1 hour, a whale chased the HOOD price with 10x leverage, with a scale of $1.12 million and an average price of $109.6. The liquidation price was set aggressively at $103.6. At the same time, the same address opened a 25x leveraged long position on BTC with a scale of $2.76 million and an average price of $60,494. It is reported that this whale had previously preferred to engage in trades related to the semiconductor sector. Against the background of today's AI infrastructure chain under pressure and the decline of the Japanese and Korean stock markets, the whale shifted its focus to an independent asset decoupled from the AI computing power narrative, bypassing sector systemic risks. On the news front, Robinhood launched its self-built Arbitrum Layer 2 chain, Robinhood Chain, officially supporting 24/7 tokenized stock trading, driving its independent surge against the broader market. The HOOD token on the Hyperliquid platform has seen a 11.2% increase in the past 24 hours, followed by another 3.7% surge in pre-market trading, now trading at $111, with a trading volume of approximately $15.92 million during the same period.

But currently, the real situation seems to be: Meta is cashing out its stock of old computing power, rather than stopping its pursuit of high-end computing power. In mid-to-late June, it was reported that Meta had signed an agreement with Crusoe, planning to obtain a total of about 1.6GW of AI computing capacity from two data centers in Texas and Missouri. At the same time, Meta's Q1 2026 still raised its full-year capital expenditure guidance to $125-145 billion. Looking at these two things together, it appears more like a reallocation of resources of different generations and purposes: continuing to buy new cards to train cutting-edge models, while old cards are used for inference in high-traffic products, hosting external models, and other application scenarios, making it acceptable to rent out a portion. However, this does not mean that purchases of the most in-demand high-end cards will slow down.